Anatomically complete spinal cord injury (SCI) transects and eliminates all functional connections across the level of the lesion, and in adults, axons fail to regrow spontaneously across such lesions. Restoring voluntary control of function will require interventions to establish new neural connections across the lesion. During the previous funding cycle of this grant, we identified a mechanism-based biological repair strategy for achieving robust regrowth of propriospinal axons across complete SCI lesions in rodents. We showed that providing three mechanisms essential for axon growth during development, (i) neuron intrinsic growth capacity, (ii) growth-supportive substrate and (iii) chemoattraction, can achieve robust regrowth of axons through and beyond anatomically complete SCI. This axon regrowth was 100-fold greater than controls, passed a full spinal segment beyond the injuries, and was able to restore significant electrophysiological conduction capacity across injuries. To achieve the spatially and temporally controlled in vivo molecular delivery required to realize this axon regrowth, we engineered biomaterial depots that enabled us to mimic certain spatiotemporal events regulating axon growth during development. In the project proposed here, we will build on this work and use our newly developed synthetic hydrogel vehicle to deliver molecules that direct the differentiation in vivo of grafted neural progenitor cells (NPC) into axon-supportive immature astroglia that repopulate non-neural lesion cores and reestablish a multicellular neural environment favorable for long term support of host propriospinal axons chemoattracted to regrow through lesions into spared neural tissue. Our hypothesis is that repopulating (and ?reneuralizing?) such non-neural lesion cores, or their cysts, with immature astroglia will promote long-term axonal maintenance and provide a favorable niche for remyelinating cells. Our objective is to develop engineering approaches that facilitate doing so. Our premise is that our hydrogel vehicles can deliver both: (i) molecules that direct the differentiation of NPC in vivo, and (ii) molecules that chemoattract host axons. Our past work and preliminary data show that NPC grafted in our hydrogel vehicles are good candidates to generate support cells for host axons as well as for host-derived oligodendrocyte progenitor cells that migrate into areas of grafted cells. We have also shown that propriospinal neurons are good targets for bridging host axons across complete SCI lesions into spared neural tissue below injuries. The work for this proposal will advance the development of mechanism- based engineering approaches to repair neural tissue after severe SCI, stroke and other CNS disorders with large focal lesions.